2 May 2018 in IEEE Journal on Emerging and Selected Topics in Circuits and Systems
Approximate computing has been considered to improve the accuracy-performance trade-off in error-tolerant applications. For many of these applications, multiplication is a key arithmetic operation. Given that approximate compressors are a key element in the design of power-efficient approximate multipliers, we first propose an initial approximate 4:2 compressor that introduces a rather large error to the output. However, the number of faulty rows in the compressor’s truth table is significantly reduced by encoding its inputs using generate and propagate signals. Based on this improved compressor, two 4×4 multipliers are designed with different accuracies and then are used as building blocks for scaling up to 16×16 and 32×32 multipliers. According to the mean relative error distance (MRED), the most accurate of the proposed 16×16 unsigned designs has a 44% smaller power-delay product (PDP) compared to other designs with comparable accuracy. The radix-4 signed Booth multiplier constructed using the proposed compressor achieves a 52% reduction in the PDP-MRED product compared to other approximate Booth multipliers with comparable accuracy. The proposed multipliers outperform other approximate designs in image sharpening and joint photographic experts group (JPEG) applications by achieving higher quality outputs with lower power consumptions. For the first time, we show the applicability and practicality of approximate multipliers in multiple-input multiple-output (MIMO) antenna communication systems with error control coding.
8 October 2019 in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
Improving the accuracy of a neural network (NN) usually requires using larger hardware that consumes more energy. However, the error tolerance of NNs and their applications allow approximate computing techniques to be applied to reduce implementation costs. Given that multiplication is the most resource-intensive and power-hungry operation in NNs, more economical approximate multipliers (AMs) can significantly reduce hardware costs. In this article, we show that using AMs can also improve the NN accuracy by introducing noise. We consider two categories of AMs: 1) deliberately designed and 2) Cartesian genetic programing (CGP)-based AMs. The exact multipliers in two representative NNs, a multilayer perceptron (MLP) and a convolutional NN (CNN), are replaced with approximate designs to evaluate their effect on the classification accuracy of the Mixed National Institute of Standards and Technology (MNIST) and Street View House Numbers (SVHN) data sets, respectively. Interestingly, up to 0.63% improvement in the classification accuracy is achieved with reductions of 71.45% and 61.55% in the energy consumption and area, respectively. Finally, the features in an AM are identified that tend to make one design outperform others with respect to NN accuracy. Those features are then used to train a predictor that indicates how well an AM is likely to work in an NN.
1 April 2021 in IEEE Transactions on Computers
Multiplication is the most resource-hungry operation in neural networks (NNs). Logarithmic multipliers (LMs) simplify multiplication to shift and addition operations and thus reduce the energy consumption. Since implementing the logarithm in a compact circuit often introduces approximation, some accuracy loss is inevitable in LMs. However, this loss is tolerated by the inherent error tolerance of NNs and their associated applications. This article proposes an improved logarithmic multiplier (ILM) that, unlike existing designs, rounds both inputs to their nearest powers of two by using a proposed nearest-one detector (NOD) circuit. Considering that the output of the NOD uses a one-hot representation, some entries in the truth table of a conventional adder cannot occur. Hence, a compact adder is designed for the reduced truth table. The 8×8 ILM achieves up to 17.48% saving in power consumption compared to a recent LM in the literature while being almost 15% more accurate. Moreover, the evaluation of the ILM for two benchmark NN workloads shows up to 21.85% reduction in energy consumption compared to the NNs implemented with other LMs. Interestingly, using the ILM increases the classification accuracy of the considered NNs by up to 1.4% compared to a NN implementation that uses exact multipliers.
9 October 2018 in IEEE Transactions on Nanotechnology
In the Internet of Things era, security concerns may require a cryptography system in every connected device. True random number generators (TRNGs) are preferred instead of pseudo-random number generators in the cryptography systems to achieve a higher level of security. For on-chip applications, we seek scalable and CMOS-compatible devices and designs for TRNGs. In this article, the stochastic behavior of the spin transfer torque magnetic tunnel junction (STT-MTJ) is utilized for the source of randomness. However, variations and correlations exist in MTJs due to fabrication limitations, so TRNG designs based on a single MTJ have to be post-processed or tracked in real time to ensure an acceptable level of randomness. Two novel designs are proposed in this article which can produce random sequences with high variation-resilience. The first design uses a parallel structure to minimize variation effects, and the second design leverages the symmetry of an MTJ-pair to take advantage of any correlations. Moreover, a universal circuit for quality improvement is proposed and it can be used with any random number generator. All of the designs are validated in a 28-nm CMOS process by Monte Carlo simulation with a compact model of the MTJ. The National Institute of Standards and Technology (NIST) statistical test suite is used to test the randomness quality of the generated sequences under the scenario of encryption keys in Transport Layer Security or Secure Sockets Layer (TLS/SSL) cryptographic protocol.
2 November 2023 in IEEE Transactions on Circuits and Systems I: Regular Papers
The increasing computational intensity of important new applications poses a challenge for their use in resource-restricted devices. Approximate computing using power-efficient arithmetic circuits is one of the emerging strategies to reach this objective. In this article, five hardware-efficient logarithmic floating-point (FP) multipliers are proposed, which all use simple operators, such as adders and multiplexers, to replace complex and more costly conventional FP multipliers. Radix-4 logarithms are used to further reduce the hardware complexity. These designs produce double-sided error distributions to mitigate error accumulation in complex computations. The proposed multipliers provide superior trade-offs between accuracy and hardware, with up to 30.8% higher accuracy than a recent logarithmic FP design or up to $68\times$ less energy than the conventional FP multiplier. Using the proposed FP logarithmic multipliers in JPEG image compression achieves higher image quality than a recent logarithmic multiplier design with up to 4.7 dB larger peak signal-to-noise ratio. For training in benchmark NN applications, the proposed FP multipliers can slightly improve the classification accuracy while achieving $4.2\times$ less energy and $2.2\times$ smaller area than the state-of-the-art design.
4 April 2018 in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
Triple-modular redundancy (TMR), which consists of three identical modules and a voting circuit, is a common architecture for soft-error tolerance. However, the original TMR suffers from two major drawbacks: the large area overhead and the vulnerability of the voter. In order to overcome these drawbacks, we propose a new complementary dual-modular redundancy (CDMR) scheme for mitigating the effect of soft errors. Inspired by the Markov random field (MRF) theory, a two-stage voting system is implemented in CDMR, including a first-stage optimal MRF structure and a second-stage high-performance merging unit. The CDMR scheme can reduce the voting circuit area by 20% while saving the area of one redundant module, achieving at least 26% error-rate reduction at an ultralow supply voltage of 0.25 V with 8.33% faster timing compared to previous voter designs.